The regulated release of peptide hormones, neural peptides, growth factors and monoamines depends on their storage inside large dense core vesicles (LDCVs) capable of regulated exocytosis. However, we still understand remarkably little about how proteins sort into this regulated secretory pathway (RSP) rather than the constitutive secretory pathway that confers the immediate release of most newly synthesized proteins. In the trans-Golgi network (TGN), proteins destined for LDCVs aggregate to form a dense core, suggesting that lumenal or possibly membrane interactions drive LDCV biogenesis, with proteins destined for other organelles removed during the subsequent process of LDCV maturation. However, we have previously identified a cytoplasmic motif required for the sorting of vesicular monoamine transporter VMAT2 into LDCVs, suggesting a role for cytosolic machinery. Mutations in this motif increase cell surface expression of the transporter, apparently by diverting it from the regulated to the constitutive pathway. Reasoning that a defect in LDCV biogenesis should phenocopy the effect of these mutations in VMAT2, we screened for increased cell surface expression of the transporter in Drosophila S2 cells, which are highly susceptible to RNAi. We find that S2 cells express an RSP and remarkably, Drosophila VMAT (dVMAT) contains the same sorting motif as the mammalian transporter, with mutations in this motif also increasing cell surface expression in S2 cells. Screening 7000 Drosophila sequences conserved to mammals by flow cytometry for increased expression of wild type dVMAT, we identified a small number of genes that affect regulated protein secretion. Focusing on the heterotetrameric adaptor protein AP-3 because two of the subunits scored positive in the screen, we found that loss of AP-3 also dysregulates secretion in mammalian cells. Although LDCVs still form in the absence of AP-3, we find that they lack the proteins such as synaptotagmin required for regulated release. In the first two aims, we will determine how AP-3 contributes to formation of the RSP by testing the hypothesis that AP-3 functions to segregate cargo destined for the RSP, and in its absence, the two secretory pathways mix. We have also found that knockdown of the AP-3-interacting protein VPS41 dysregulates protein secretion, and in the third aim, will test the hypothesis that VPS41 functions as a coat protein for the AP-3 adaptor. We will also extend the analysis to mice lacking AP-3 and VPS41. The results will provide a foundation for future work on the molecular mechanisms involved in LDCV formation and the consequences for physiology, development and disease.